Collaborative filtering through SVD-based and hierarchical nonlinear PCA
ICANN'10 Proceedings of the 20th international conference on Artificial neural networks: Part I
| Hi-index | 0.01 |
Recommendation systems can help people to find interesting things and they are widely used in our life with the development of the Internet. Collaborative filtering technique has been proved to be one of the most successful techniques in recommendation systems in recent years. Poor quality is one major challenge in collaborative filtering recommender systems. Sparsity of source data set is the major reason causing the poor quality. Aiming at the problem of data sparsity for collaborative filtering, a new personalized recommendation approach based on BP neural networks and item based collaborative filtering is presented. This method uses the BP neural networks to fill the vacant ratings where necessary and uses item based collaborative filtering to form nearest neighborhood, and then generates recommendations. The experiment results argue that the algorithm efficiently improves sparsity of rating data, and promises to make recommendations more accurately than conventional collaborative filtering.